Skip to content

[Docker] Install cudnn==9.16 for cuda 13 image to avoid check error#17668

Merged
Fridge003 merged 1 commit intomainfrom
baizhou/patch-5
Jan 24, 2026
Merged

[Docker] Install cudnn==9.16 for cuda 13 image to avoid check error#17668
Fridge003 merged 1 commit intomainfrom
baizhou/patch-5

Conversation

@Fridge003
Copy link
Copy Markdown
Collaborator

Motivation

  File "/sgl-workspace/sglang/python/sglang/srt/server_args.py", line 4975, in check_torch_2_9_1_cudnn_compatibility
    raise RuntimeError(msg)
RuntimeError: CRITICAL WARNING: PyTorch 2.9.1 & CuDNN Compatibility Issue Detected

Modifications

Accuracy Tests

Benchmarking and Profiling

Checklist

Review Process

  1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
  2. Get approvals from CODEOWNERS and other reviewers.
  3. Trigger CI tests with comments or contact authorized users to do so.
    • /tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci
  4. After green CI and required approvals, ask Merge Oncalls to merge.

@gemini-code-assist
Copy link
Copy Markdown
Contributor

Summary of Changes

Hello @Fridge003, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical runtime error related to PyTorch 2.9.1 and CuDNN compatibility when using CUDA 13. By explicitly installing the correct CuDNN version in the Docker image for CUDA 13, the change ensures that applications can run without encountering the reported compatibility issue.

Highlights

  • CUDA 13 CuDNN Installation: Added nvidia-cudnn-cu13==9.16.0.29 to the Dockerfile for CUDA 13 environments to resolve a compatibility issue with PyTorch 2.9.1 and CuDNN.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request correctly addresses a runtime error related to PyTorch and CuDNN compatibility for CUDA 13 by adding the installation of nvidia-cudnn-cu13. The change is straightforward and necessary. I've included a suggestion to group the pip install commands for better maintainability and a minor build performance improvement, consistent with other parts of the Dockerfile.

Comment thread docker/Dockerfile
Comment on lines 311 to 315
python3 -m pip install nvidia-nccl-cu13==2.28.3 --force-reinstall --no-deps ; \
python3 -m pip install nvidia-cudnn-cu13==9.16.0.29 --force-reinstall --no-deps ; \
python3 -m pip install nvidia-cublas==13.1.0.3 --force-reinstall --no-deps ; \
python3 -m pip install nixl-cu13 --no-deps ; \
python3 -m pip install cuda-python==13.1.1 ; \
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

For better maintainability and to slightly improve build performance by reducing the number of pip processes, consider grouping these pip install commands based on their options. This is also more consistent with how packages are installed elsewhere in this Dockerfile.

    python3 -m pip install --force-reinstall --no-deps \
        nvidia-nccl-cu13==2.28.3 \
        nvidia-cudnn-cu13==9.16.0.29 \
        nvidia-cublas==13.1.0.3 ; \
    python3 -m pip install nixl-cu13 --no-deps ; \
    python3 -m pip install cuda-python==13.1.1 ; \

@Fridge003 Fridge003 merged commit 0dfe46d into main Jan 24, 2026
51 checks passed
@Fridge003 Fridge003 deleted the baizhou/patch-5 branch January 24, 2026 03:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant